Comment 31 for bug 1501781

https://gist.github.com/jfragoulis/9914900 gives an example of how to customize Pg's unaccent(), but for the sake of consistency, I'm wondering if we might be better off taking the performance hit and and writing evergreen.unaccent_and_squash() in PL/Perl so that it can use Text::Unaccent::PurePerl.